********************************************************
***  PLOTTING TOOLS, REPLICATION FILES AND DATA FOR  ***
***                                                  ***
*** "Empirical evidence on in
flation expectations in ***
***         the new Keynesian Phillips curve"        ***
***                                                  ***
*** by Sophocles Mavroeidis, Mikkel Plagborg-Moller  ***
***              and James H. Stock                  ***
***                                                  ***
***                  05/27/2013                      ***
********************************************************


*** CONTACT
If you have questions or comments, please email:

- Sophocles Mavroeidis:   sophocles.mavroeidis@economics.ox.ac.uk
- Mikkel Plagborg-Moller: plagborg@fas.harvard.edu
- James Stock:            james_stock@harvard.edu


*** QUICKSTART

* To use the Matlab plotting tools:
Open the folder empirics/matlab/ and execute either:
- plot_estimates.m (point estimates)
- plot_confsets.m (robust confidence sets)
See the documentation file documentation/plotting_tools.pdf
for further information.

* To access our data:
Open empirics/stata/nkpc.dta in Stata. Please refer to
documentation/dbdoc.pdf for details.


*** CONTENTS

** Documentation

- dbdoc.pdf: documents the Stata database nkpc.dta
- plotting_tools: documents the Matlab plotting tools
- rt_ulc_ls.pdf: documents the real-time labor share data set
- simulations.pdf: documents the Monte Carlo simulations and
  provides results for additional DGPs

** Empirics

* Matlab
- first_stage_plot.m: plots kernel density estimate of first-stage
  F statistics
- GGLS.txt: estimation results using Gali, Gertler and
  Lopez-Salido (2001) vintage data
- lit_estims.m: plots labor share estimates from the literature
- load_data.m: loads estimation data produced in Ox
- plot_all.m: produces plots will all point estimates
- plot_confsets.m: robust confidence set plotting tool
- plot_estimates.m: point estimate plotting tool
- folder surveys_vs_RE/: plots robust confidence sets for NKPC
  specifications with both next-period inflation and survey
  forecasts
- other files: please refer to documentation/plotting_tools.pdf

* rt_laborshare
- please refer to documentation/rt_ulc_ls.pdf

* Stata
- create.do: code used to create the Stata database, lists details
  about the various data series
- doc_trend_inf.txt: abbreviations used in trend_inf.dta file
- detrend.do: code for detrending labor share or output
- nkpc.dta: main database, please refer to documentation/dbdoc.pdf
- rt_rgdp_20120225.csv: real-time output data
- series_20120225.csv: raw data series
- trend_inf.dta: additional database with Stock-Watson trend
  inflation


* Ox

- Estimate_all.ox: computes the point estimates that are used in
  the Matlab plotting tool plot_estimates.m

- Compute_confsets.ox: computes the confidence sets that are used

  in the Matlab plotting tool plot_confsets.m

- Compute_confsets_Survey_vs_RE.ox: computes the confidence sets
  for the results in section 5.4 of the paper

- nkpc_20121219.bn7 and nkpc_20121219.in7: Ox database
- folder gmm/: contains GMM routines and tests of weak
  identification


** Simulations

* aim_fuhrer
- Jeff Fuhrer's AIM code (Anderson and Moore, 1985). The files
  whose names end in "_mpm" have been lightly edited. Source:
  http://www.bos.frb.org/economic/econbios/fuhrer/matlab.htm

* dynare
- drawplot.m: plots impulse responses
- irf_plot.m: calculates impulse responses for a range of
  parameter values
- nk.mod: Dynare model file for the 3-equation new Keynesian model
  with a hybrid NKPC

* mc
- disp_output_var.m: displays simulation results
- drawdens.m: plots estimated sampling density of estimators
- gmm.m: linear GMM estimation
- md.m: VAR-MD estimation
- ml_con.m: constraint for VAR-ML estimation
- ml_kurmann.m: VAR-ML a la Kurmann (2007)
- nws.m: Newey-West (1987) HAC estimation
- observ1.m: AIM calculations used in specifying DGPs
- run_mc_var.m: main executable for Monte Carlo simulations of
  OLS, GIV, VAR-GMM, VAR-MD and VAR-ML estimators; note that names
  of DGPs are those used in documentation/simulations.pdf!
